Author Topic: Need advice on designing a stepper motor application that can be commercialized  (Read 4207 times)

0 Members and 1 Guest are viewing this topic.

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Hi,

I have a stepper/servo motor application that I want to commercialize, and here are the requirements:

1. The stepper motor will be rotating back and forth like a pendalum (180 degree total travel), with a period of roughly 2-3 sec.
2. The torque shall be powerful enough to drive a cassette player.
3. I really prefer not to use Li-ion as it has a dangerous reputation and I want to commercialize something with less risk.
4. I hope the battery can be recharged via a USB cable, rather than using alkaline batteries.
5. Operation should be quiet.

So far, I only found the 28 BYJ48 motor to have the size, cost, availability, torque, and quietness attribute that I need. However, it requires voltage of at least 4V. The drive circuit is simple. Lots of off the shelf drivers uses ULN2003, but I found a more efficient way using MOSFET.

The simplest way would be to use 4 AA batteries to power it. But I hate to have to swap batteries all the time. How do I design the device so that it can be charged via USB like a cell phone?

Maybe I can use a single AA battery, but that'll require a boost circuit which is hard to design. Moreover, the charging circuit is kinda out of my league now too.

Should I just go with the simple approach (4 AA Nimh pack) and commercialize it first? (have a minimal working product), or try to go fancy and go with USB charging? Unless there are off the shelf circuits that'll do the charging and boost the voltage, it'll be too much of a risk from a time and design stand point. Last thing I want is have the device start a fire.

I researched boost circuit, and the best I can do is to buy an IC chip and I still have to design around it. While I'm always up for a challenge, I don't feel it's the best approach from a business stand point.

What is your suggestion?

Thanks
 

Offline Nitrousoxide

  • Regular Contributor
  • *
  • Posts: 156
  • Country: au
To get the energy density required and to satisfy the recharging requirement, you will have to use Li-Po/Li-ion batteries.

If you want simplicity, you can get whole boost modules that even include the inductor (they are usually more expensive). However, you can just go for a boost converter (integrated MOSFET). Most boost converters will have a recommended or example layout in the datasheet, copy that and it will be fine. I personally did that for my first PCB and ran into no issues at all.

Some controllers can even user micro stepping to reduce the stepper noise
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11534
  • Country: my
  • reassessing directives...
Maybe I can use a single AA battery, but that'll require a boost circuit which is hard to design. Moreover, the charging circuit is kinda out of my league now too.
you want to commercialize a thing and you are not up to the task yet? ironic. maybe you can start learning and experimenting step by step. first charging circuit, and then boost circuit etc etc...

How do I design the device so that it can be charged via USB like a cell phone?
your best guru will be the existing commercialized products, licensed, tested and verified by real engineers and standard organizations. what battery cell phones use? right, lion or lipo (worst safety). do you see most products around use NiMH or SLA (safer)? no (or very seldomly seen), why? since your research contradicts with what happening around you, there must be something flawed in your finding (or methodology or mentally) imho. just because of few isolated incidents and you are not going to use it is like banning petrol or hydrogen car from the road because of some explosions petrol or hydrogen related.

I researched boost circuit, and the best I can do is to buy an IC chip and I still have to design around it. While I'm always up for a challenge, I don't feel it's the best approach from a business stand point.
can you point out which product (consumers grade) design their own boost circuit from raw transistors?

there are many possibilities in the advice, use lion/lipo/lifepo4 (because it norm today), or use NiMH or SLA (i'm sure you dont want that) and then googling for the charge circuit and read books on how each type of battery should be properly charged. i know this is beginner section but instead of suggesting on one particular and narrow (minded) possibility, i prefer provoking your mind so hopefully it will directed in the right way as many proven commercial companies do. hope it helps.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Li-Ion is the way everything has gone, if you build a commercial product with NiCd or NiMH it's going to look archaic. There's no reason to be afraid of Li-ion as long as the charging circuitry is designed sensibly, the high current LiPo packs used in RC models and such can be dangerous but Li-ion are used in literally billions of mobile devices and laptops with very few incidents. NiCd, NiMH and lead acid batteries can also start fires, though usually in the wiring. Any time you have stored energy you can have a fire.
 

Offline stian

  • Contributor
  • Posts: 14
  • Country: no
Hi,

I have a stepper/servo motor application that I want to commercialize, and here are the requirements:

1. The stepper motor will be rotating back and forth like a pendalum (180 degree total travel), with a period of roughly 2-3 sec.
What is your suggestion?
Thanks

Hi.

My 1st question back to you is. who in their right mind do you think would buy a software that does this?
My answer is no one i think. and there is no point in that.

And more so. the 28 BYJ48 motor is rather a toy stepper, and with the ULN2003 they are rather slow if you ask me.
Also the ULN2003 driver + the 28 BYJ48 stepper does not work on 4v. you need 10V+ (12v preferable) on the steppers for them to have any torque.
I myself is using these 28 BYJ48 motor and the ULN2003 driver. "i bough them on ebay cheap :p" in controlling a usb microscope im 3d printing out.
im using a arduino uno "Atmega328" with code i written in C "not C++" to control the 4 stepper motors over Usb/serial.

cant really say i would reccoment the 28 BYJ48 motor and the ULN2003 driver
my project involving them. https://i.gyazo.com/814da14d11abdafc40e51d200ef6c370.jpg

You should maybe look for a opensource application that already exists that uses Gcode or something and fork and customize it.
also is your application only going to support one type of stepper? there are many types of stepper motors you know. Unipolar, Bipolar, 4-Wire Motor, 5-Wire Motor.. and so on.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
A single Li-ion cell will provide sufficient voltage without a boost converter and there are single IC charge management solutions. There are much better stepper drivers than the ULN series or discrete mosfets, TI makes some that can do current regulation, microstepping and such that are not hard to use. I suggest looking into those.
 

Offline Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Why reinvent the wheel? There are probably thousands of different USB battery bank packs out there, which are charged via USB and provide power via USB. If you design your project to run directly from USB, then the power aspect is just a matter of buying one with sufficient capacity for your requirements.

My suggestion is don't get hung up on the final power source right now. Instead use whatever 5V supply is handy and get your prototype working first.
 

Online Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 3321
  • Country: nl
What kind of quantity do you expect to sell of this product?

It seems strange to me you want to build a commercial product, but apparently do not have the skills to design it. Maybe you should find a partner for the design.

The USB power banks (should) have built in protection for a Li-Po battery. Because of the vast numbers of these things built, special chips are made which combine over and under voltage protection and step-up to 5V in a single IC.

Be carefull with the 28Cxxxx stepper motors. These are built by multiple different manufacturers and with different gear ratio's (1:16 and 1:64 are both common) and voltages. (6V and 12V is common).
I've bought some of these with the same voltage rating, and they had a pretty big deviation in DC coil resistance.
Look closely at pictures of a bunch of these on Ali / Ebay / Etc. You will see small differences between them.

With battery operation, power consumption is always a big issue. Stepper motors are bad at this. They consume the most energy when doing nothing, if you constantly apply voltage to the coils. You could also increase efficiency by opening them and wiring them as bipolar motors and use a proper dirver chip for them.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
This is where the specialized driver ICs help out, some of them can do PWM current limiting which greatly reduces the holding current. They're still not great for battery powered applications though.
 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Li-Ion is the way everything has gone, if you build a commercial product with NiCd or NiMH it's going to look archaic. There's no reason to be afraid of Li-ion as long as the charging circuitry is designed sensibly, the high current LiPo packs used in RC models and such can be dangerous but Li-ion are used in literally billions of mobile devices and laptops with very few incidents. NiCd, NiMH and lead acid batteries can also start fires, though usually in the wiring. Any time you have stored energy you can have a fire.

I know Li-Ion is used in most consumer devices now. But a few years ago there was Sony laptop fires and also fire with the B787 batteries. Not to mention that Li-Ion cannot be checked on flights makes one think they are inherently dangerous. If even global companies can mess up with them, then that makes me think there is always a real risk. I have not heard of any incidents with NiMh. Correct me if I'm wrong.


 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
Why reinvent the wheel? There are probably thousands of different USB battery bank packs out there, which are charged via USB and provide power via USB. If you design your project to run directly from USB, then the power aspect is just a matter of buying one with sufficient capacity for your requirements.


I really like this idea, thanks!
 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us

And more so. the 28 BYJ48 motor is rather a toy stepper, and with the ULN2003 they are rather slow if you ask me.
Also the ULN2003 driver + the 28 BYJ48 stepper does not work on 4v. you need 10V+ (12v preferable) on the steppers for them to have any torque.
I myself is using these 28 BYJ48 motor and the ULN2003 driver. "i bough them on ebay cheap :p" in controlling a usb microscope im 3d printing out.
i
also is your application only going to support one type of stepper? there are many types of stepper motors you know. Unipolar, Bipolar, 4-Wire Motor, 5-Wire Motor.. and so on.

I plan to use the TI DRV777 to drive the 28 BYJ48. Please tell me what you think:
http://www.ti.com/lit/ds/symlink/drv777.pdf

They seem a lot more efficient than the ULN2003.

My application has really no requirement on the stepper. It doesn't even need to be a stepper. It's just that the 28 byj48 motor is the only one I've seen that's cheap enough, the right size, the right torque, and mostly importantly, quiet.  Most hobby servos makes a audible whine when they move. I also bought some really tiny 10mm or 16mm stepper motors and those don't seem to have enough torque, and didn't come in any common size. The 28 byj48 on the other hand is widely available and cheap.

If you know of any other options, pls let me know. My application is slow speed, and quietness is important. I wonder what kind of motor is used for cassette players because they seem to have the right speed, noise level, and torque for my application.

Thanks
 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
What kind of quantity do you expect to sell of this product?

It seems strange to me you want to build a commercial product, but apparently do not have the skills to design it. Maybe you should find a partner for the design.

The USB power banks (should) have built in protection for a Li-Po battery. Because of the vast numbers of these things built, special chips are made which combine over and under voltage protection and step-up to 5V in a single IC.

Be carefull with the 28Cxxxx stepper motors. These are built by multiple different manufacturers and with different gear ratio's (1:16 and 1:64 are both common) and voltages. (6V and 12V is common).
I've bought some of these with the same voltage rating, and they had a pretty big deviation in DC coil resistance.
Look closely at pictures of a bunch of these on Ali / Ebay / Etc. You will see small differences between them.

With battery operation, power consumption is always a big issue. Stepper motors are bad at this. They consume the most energy when doing nothing, if you constantly apply voltage to the coils. You could also increase efficiency by opening them and wiring them as bipolar motors and use a proper dirver chip for them.

Quantity? 10000 to 20000 hopefully

This project is slightly beyond my skill at the moment, but if I use the power bank idea like you mentioned, then I think the rest is doable. I really like the power bank idea. I don't even need to worry about the charging or the boost converter. I can just design my device to interface with any USB power bank and the customer can purchase the power bank on their own. Some of those small tubular USB battery bank actually looks pretty nice with my design.

As for the stepper motor being inefficient, so far I have not found a better solution in terms of quietness, torque, size, cost/availability than the 28 byj48. My post above this provides more detail.

Thanks
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
If possible I buy products that use 4 AA cells (or maybe 4 AAA cells) so that I can stick 4 Eneloops in there. Advantage: I have a stock of Eneloops I can swap between products, I can simply replace them with fresh ones when they run down, and if necessary I can stick alkaline cells in there. I can buy alkaline cells at 4/$1 at Dollar Tree, which is cheap enough to be disposable.

I try to avoid products that use Li-Ion unless it is a replaceable battery of a standard size, since the life of the battery will be the life of the product. Given that the product will last forever and the battery will not, Li-ion is a bad idea.

But I am not your typical consumer...

(Your typical consumer will buy the product, get bored of it within a year, and dispose of it.)
 

Offline Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
I wonder what kind of motor is used for cassette players because they seem to have the right speed, noise level, and torque for my application.

Google "cassette player motor" if you really want to know.

It's difficult to recommend anything without knowing the precision requirements of your project.
 

Offline SL4P

  • Super Contributor
  • ***
  • Posts: 2318
  • Country: au
  • There's more value if you figure it out yourself!
My question is more fundamental.
What’s the fixation with a stepper?
A properly controlled DC motor on an h-bridge will be infinitely quieter, more controllable, draw less power overall, and smoother when implementing a ‘pendulum’ motion effect.
Don't ask a question if you aren't willing to listen to the answer.
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11534
  • Country: my
  • reassessing directives...
Quantity? 10000 to 20000 hopefully
good luck!
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
My question is more fundamental.
What’s the fixation with a stepper?
A properly controlled DC motor on an h-bridge will be infinitely quieter, more controllable, draw less power overall, and smoother when implementing a ‘pendulum’ motion effect.

You are right. The stepper is not necessary. It's just that I've found a quiet one that suits my need.

My previous experience with DC motors tell me they are loud. The ones I've had experience are the following type:
But then again, they are cheap, hobby types and spins at thousands of rpm. It' s hard to be quiet at that speed. To get it slow enough I need, I thought about using a gearbox which would add to the noise.

I have not tried using H bridge. I heard you can use PWM to slow it down but the speed I need are so slow (like a cassette player), I'm not sure if I can electronically control it to be that slow without a gearbox. Please advise. Thanks
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11534
  • Country: my
  • reassessing directives...
My previous experience with DC motors tell me they are loud...
at that speed. To get it slow enough I need, I thought about using a gearbox which would add to the noise.
if you can control the power by PWM to the required speed, you'll get alot quieter motor and no gearbox required.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
With the right motor you could directly drive a pendulum, controlling the current so the motor is not spinning a full revolution, just applying torque to the pendulum. Stepper motors are great when you want precise absolute position control but it doesn't sound like you need that here.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
if you can control the power by PWM to the required speed, you'll get alot quieter motor and no gearbox required.

Depends on the PWM frequency. PWM will typically make motors buzz or whistle.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
It's certainly not hard to make the frequency high enough to be inaudible, especially with low power stuff where switching losses aren't so important.
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11534
  • Country: my
  • reassessing directives...
Well i mean smps actually to be more robust and less whining due to less volt ripple, and yes use freq hi enough.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline engineheatTopic starter

  • Frequent Contributor
  • **
  • Posts: 266
  • Country: us
My previous experience with DC motors tell me they are loud...
at that speed. To get it slow enough I need, I thought about using a gearbox which would add to the noise.
if you can control the power by PWM to the required speed, you'll get alot quieter motor and no gearbox required.


Is there a limit to how slow PWM can make a DC motor turn? For example, a motor that spins at 5000 rpm under light load, can you really make it spin at 50 rpm with PWM?

How will the load affect the rpm? I don't want the motor to spin at different speeds at different load. The stepper has the benefit in that its noise/speed is pretty much constant whether there is load or not.

Thanks
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
You can make a motor spin at 0 RPM if you like, just applying torque to an object but not actually spinning it.

Speed will of course vary based on load, unless you have an encoder or other method of positional feedback. It's really hard to say without knowing more about the design you are working on.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf